© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 15556
    Prev
    Next
    INDEX
    Explanations

    imperfections or early arrival

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is detecting relatively rare or low‐frequency content words (uncommon vocabulary items).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ों
    0.87
    ੀ
    0.73
    POT
    0.72
    Ра
    0.71
     ହ
    0.71
    s
    0.71
    د
    0.70
    PEN
    0.68
    ER
    0.66
    声道
    0.66
    POSITIVE LOGITS
    hető
    0.87
    igenschaften
    0.81
    واں
    0.80
     પછી
    0.79
    tery
    0.79
     फिर
    0.78
     termasuk
    0.78
     terecht
    0.78
     پھر
    0.77
    轳
    0.77
    Activations Density 0.001%

    No Known Activations